Veterans Administration (VA) Million Veteran Program (MVP) Summary Results from Omics Studies

MVP is an ongoing prospective cohort study and mega-biobank in the Department of Veterans Affairs Healthcare System designed to study genetic influences on health and disease among veterans.]]>
Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with inverse normal transformed HDL cholesterol levels stratified by ethnicity followed by a meta-analysis of results across all three groups.Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with inverse normal transformed LDL cholesterol levels stratified by ethnicity followed by a meta-analysis of results across all three groups.Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with inverse normal transformed total cholesterol levels stratified by ethnicity followed by a meta-analysis of results across all three groups.Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with inverse normal transformed triglyceride levels (after natural log transformation) stratified by ethnicity followed by a meta-analysis of results across all three groups.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for diastolic blood pressure (DBP) using samples from Million Veteran Program transethnic meta-analysis.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for pulse pressure (PP) using samples from Million Veteran Program transethnic meta-analysis.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for systolic blood pressure (SBP) using samples from Million Veteran Program transethnic meta-analysis.Association testing was performed in up to 31,307 PAD cases and 211,753 controls in the Million Veteran Program. Pariticpants of European (white), African (black), and Hispanic ancestry were stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups.Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups.Association testing was performed in up to 297,626 white (European ancestry), black (African ancestry), and Hispanic MVP participants with blood lipids stratified by ethnicity followed by a meta-analysis of results across all three groups.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for estimated glomerular filtration rate (eGFR) using samples from Million Veteran Program transethnic meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis. Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in up to 274,424 European Americans (EA), African Americans (AA), Latino Americans (LA), East Asian Americans (EAA) and South Asian Americans (SAA) in MVP for alcohol consumption (AUDIT-C, N=272,842) and alcohol use disorder (N=274,391) stratified by ethnicity followed by a trans-population meta-analysis.Association testing was performed in 17,029 black (African ancestry) MVP participants with maximum habitual alcohol intake. Association testing was performed in 126,936 white (European ancestry) MVP participants with maximum habitual alcohol intake. Association testing was performed on PTSD reexperiencing symptoms in the Million Veteran Program. Particpants of white (European ancestry) and black (African ancestry)were stratified by ethnicity followed by a meta-analysis. Association testing was performed on PTSD reexperiencing symptoms in the Million Veteran Program. Particpants of white (European ancestry) and black (African ancestry) were stratified by ethnicity followed by a meta-analysis.Association testing was performed on PTSD reexperiencing symptoms in the Million Veteran Program. Particpants of white (European ancestry) and black (African ancestry)were stratified by ethnicity followed by a meta-analysis. We performed a African-American restricted meta-analysis including 24,646 type 2 diabetes cases and 31,446 controls in the Million Veteran Program (MVP) and Penn Medicine Biobank. Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. Only participants of African American origin were selected for this subgroup analysis (n = 53,445). The average age at study enrollment was 61.7 years, and prevalence of T2D was 43.6%. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, and 10 principal components of genetic ancestry. Summary statistics available from the Penn Medicine Biobank were obtained for meta-analysis. All cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel. Within each study a logistic regression model was used where T2D was set as the dependent variable, imputed SNP as the independent variable, and analyses were adjusted for age, gender, and the top ten principal components of genetic ancestry.We performed a Asian-specific meta-analysis including 46,511 type 2 diabetes cases and 169,776 controls in the Million Veteran Program (MVP), Biobank Japan, and Pakistan Genomic Resource. Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. The Asian MVP participants (n = 19,445) are predominantly male (93.9%) with an average age at study enrollment of 56.1 years. The prevalence of T2D was 35.5% 36.4% for Asians. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, and 10 principal components of genetic ancestry. Summary statistics available from previous reports in Biobank Japan and Pakistan Genomic Resource were obtained for meta-analysis. All cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel. Within each study a logistic regression model was used where T2D was set as the dependent variable, imputed SNP as the independent variable, and analyses were adjusted for age, gender, and the top ten principal components of genetic ancestry.We performed a European meta-analysis including 148,726 type 2 diabetes cases and 965,732 controls in the Million Veteran Program (MVP) and other biobanks. Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. MVP participants (n=1,114,458) are predominantly male subjects (92.8%) with an average age at study enrollment ranged of 68.2 years. The prevalence of T2D was 35.5%. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 0.1% using PLINK2a software. Covariates included age, gender, and 10 principal components of genetic ancestry. Summary statistics available from previous reports in the DIAMANTE Consortium, Penn Medicine Biobank, Malmo Diet and Cancer Study, and MedStar/PennCath studies were obtained for meta-analysis. All cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel with exception of the DIAMANTE consortium where genotype calls were imputed using the Haplotype Reference Consortium reference panel. Within each study a logistic regression model was used where T2D was set as the dependent variable, imputed SNP as the independent variable, and analyses were adjusted for age, gender, and the top ten principal components of genetic ancestry.We performed a Hispanic-restricted single variant analysis including 8,616 type 2 diabetes cases and 11,829 controls in the Million Veteran Program (MVP). Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. Hispanic MVP participants (n = 20,445) are predominantly male (91.2%), with an average age at study enrollment of 59.1, and the prevalence of T2D was 42.1%. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, and 10 principal components of genetic ancestry. The MVP genotyped cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8. We performed a multi-ethnic meta-analysis including 228,499 type 2 diabetes cases and 1,178,783 controls in the Million Veteran Program (MVP) and other biobanks. Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. MVP participants (n = 273,409) are comprised predominantly of male subjects (91.6%) and were classified as Europeans (72.1%), African Americans (19.5%), Hispanics (7.5%), and Asians (0.9%). The average age at study enrollment ranged from 56.1 for Asian to 68.2 for European participants. The prevalence of T2D was 35.5% for Europeans, 36.4% for Asians, 42.1% for Hispanics, and 43.6% for African Americans. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 0.1% in Europeans, and MAF > 1% in African Americans, Hispanics and Asians using PLINK2a software. Covariates included age, gender, and 10 principal components of genetic ancestry. Summary statistics available from previous reports in the DIAMANTE Consortium, Biobank Japan, Penn Medicine Biobank, Malmo Diet and Cancer Study, MedStar/PennCath studies, and the Pakistani Genomic Resource were obtained for meta-analysis. All cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel with exception of the DIAMANTE consortium where genotype calls were imputed using the Haplotype Reference Consortium reference panel. Only SNPs with ancestry-specific MAF > 1% in these studies were used. Within each study a logistic regression model was used where T2D was set as the dependent variable, imputed SNP as the independent variable, and analyses were adjusted for age, gender, and the top ten principal components of genetic ancestry.Meta-analysis for problematic alcohol use (PAU) combining MVP phase1, MVP phase2, PGC, and UK Biobank.Meta-analysis for alcohol use disorder (AUD) in MVP phase1 and MVP phase2.Meta-analysis for alcohol use disorder (AUD) in MVP phase1, MVP phase2, and PGC.Genome-wide association analysis for alcohol use disorder (AUD) in MVP phase2.Genome-wide association analysis for opioid use disorder (OUD) in European American participants from MVP part 1.Genome-wide association analysis for opioid use disorder (OUD) in European American participants from MVP part 2.Genome-wide association analysis for opioid use disorder (OUD) in European American participants from MVP parts 1 and 2.Genome-wide association analysis for opioid use disorder (OUD) in African American participants from MVP part 1.Genome-wide association analysis for opioid use disorder (OUD) in African American participants from MVP part 2.Genome-wide association analysis for opioid use disorder (OUD) in African American participants from MVP parts 1 and 2.Association testing was performed in up to 5,373 PAD cases and 42,485 controls of African (black) ancestry in the Million Veteran Program.Association testing was performed in up to 24,009 PAD cases and 150,983 controls of European (white) ancestry in the Million Veteran Program.Association testing was performed in up to 1,925 PAD cases and 18,285 controls of Hispanic ancestry in the Million Veteran Program.Association testing was performed in up to 11,844 VTE cases (8,929 white, 2,261 black, 654 Hispanic) and 211,753 controls. Participants of European (white), African (black), and Hispanic ancestry were stratified by ethnicity followed by a meta-analysis of results across all three groups. Association testing was performed in up to 2,261 VTE cases and 49,400 controls of African (black) ancestry. Association testing was performed in up to 8,929 VTE cases and 181,337 controls of European (white) ancestry. Association testing was performed in up to 654 VTE cases and 21,214 controls of Hispanic ancestry. Association testing was performed on GAD-2 anxiety symptoms in the Million Veteran Program. 24,448 particpants of African ancestry were included in this analysis. Association testing was performed on GAD-2 anxiety symptoms in the Million Veteran Program. 175,163 particpants of white (European ancestry) were included in this analysis. Genome-wide association analysis for opioid use disorder (OUD) in European American participants from MVP, Yale-Penn, and SAGE.Genome-wide association analysis for opioid use disorder (OUD) in African American participants from MVP and Yale-Penn.Genome-wide association analysis for opioid use disorder (OUD) in MVP, Yale-Penn, and SAGE.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for diastolic blood pressure (DBP) using White samples from Million Veteran Program meta-analysisContains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for pulse pressure using White samples from Million Veteran Program meta-analysisContains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for systolic blood pressure (SBP) using White samples from Million Veteran Program meta-analysisContains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for estimated glomerular filtration rate (eGFR) using samples from Million Veteran Program Black-only meta-analysis.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for estimated glomerular filtration rate (eGFR) using samples from Million Veteran Program White-only meta-analysisSmoking data from 2000-2015 were obtained from the Veteran Healthcare Administration Corporate Data Warehouse. Mapping strategies were created to classify these responses into never, past, and current smoking status. We used joint trajectory modeling to sort each participant's smoking values (current, past, never) into clusters and estimated distinct trajectories. We used age as the time scale to account for possible decreases in smoking with age. The procedure calculated each individual's probability of belonging to each trajectory and assigned the individual to the trajectory with the highest probability of membership (mostly current smoking, mixed smoking and non-smoking, mostly never smoking). Applying the phenotype definition, we identified 13,511 mostly current smokers, 23,605 mixed smokers, and 17,751 mostly never smokers in African Americans.Smoking data from 2000-2015 were obtained from the Veteran Healthcare Administration Corporate Data Warehouse. Mapping strategies were created to classify these responses into never, past, and current smoking status. We used joint trajectory modeling to sort each participant's smoking values (current, past, never) into clusters and estimated distinct trajectories. We used age as the time scale to account for possible decreases in smoking with age. The procedure calculated each individual's probability of belonging to each trajectory and assigned the individual to the trajectory with the highest probability of membership (mostly current smoking, mixed smoking and non-smoking, mostly never smoking). Applying the phenotype definition, we identified 2,920 mostly current smokers, 11,221 mixed smokers, and 7,195 mostly never smokers in Hispanic Americans.Smoking data from 2000-2015 were obtained from the Veteran Healthcare Administration Corporate Data Warehouse. Mapping strategies were created to classify these responses into never, past, and current smoking status. We used joint trajectory modeling to sort each participant's smoking values (current, past, never) into clusters and estimated distinct trajectories. We used age as the time scale to account for possible decreases in smoking with age. The procedure calculated each individual's probability of belonging to each trajectory and assigned the individual to the trajectory with the highest probability of membership (mostly current smoking, mixed smoking and non-smoking, mostly never smoking). Applying the phenotype definition, we identified 40,456 mostly current smokers, 110,403 mixed smokers, and 59,056 mostly never smokers in European Americans.Cigarettes per day was based on responses to the MVP baseline survey questionnaire. The subjects were asked, "Do you currently smoke cigarettes?", If yes, then "How many cigarettes do you smoke per day now?". If no, then "Over the entire time you smoked, on average, how many cigarettes did you smoke per day?". The responses were on a scale from 1 to 5, corresponding to 1) less than a half pack, 2) a half pack, 3) 1 pack, 4) 2 packs, and 5) more than 2 packs. We identified 17,014 and 77,515 individuals for CPD current and CPD past with non-missing responses, respectively.Cigarettes per day was based on responses to the MVP baseline survey questionnaire. The subjects were asked, "Do you currently smoke cigarettes?", If yes, then "How many cigarettes do you smoke per day now?". If no, then "Over the entire time you smoked, on average, how many cigarettes did you smoke per day?". The responses were on a scale from 1 to 5, corresponding to 1) less than a half pack, 2) a half pack, 3) 1 pack, 4) 2 packs, and 5) more than 2 packs. We identified 17,014 and 77,515 individuals for CPD current and CPD past with non-missing responses, respectively.Smoking data from 2000-2015 were obtained from the Veteran Healthcare Administration Corporate Data Warehouse. Mapping strategies were created to classify these responses into never, past, and current smoking status. Using all available electronic medical record-based smoking observations available, we used the most common (modal) value for smoking status assessment. We identified 72,729 never smokers, 71,002 past smokers, and 66,184 current smokers in the European American sample. We further contrasted ever smoked (past or current smokers) with never smoked (non-smokers) to study smoking initiation behavior. Smoking data from 2000-2015 were obtained from the Veteran Healthcare Administration Corporate Data Warehouse. Mapping strategies were created to classify these responses into never, past, and current smoking status. Using all available electronic medical record-based smoking observations available, we used the most common (modal) value for smoking status assessment. We identified 72,729 never smokers, 71,002 past smokers, and 66,184 current smokers in the European American sample. We used the modal value of smoking status classification and contrasted current with past to reflect smoking cessation phenotype.Smoking initiation was defined as a contrast between ever smoked (past or current smokers) with never smoked (non-smokers).Smoking cessation was defined as a contrast between current smokers with past smokers.We performed a African American-ancestry single variant analysis including 23,305 type 2 diabetes cases and 30,140 controls in the Million Veteran Program (MVP). Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. European MVP participants are predominantly male (87.2%), with an average age at study enrollment of 61.7 years, and the prevalence of T2D was 43.6%. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, and 10 principal components of African American genetic ancestry. The MVP genotyped cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8.We performed a Asian-ancestry single variant analysis including 893 type 2 diabetes cases and 1,560 controls in the Million Veteran Program (MVP). Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. European MVP participants are predominantly male (93.9%), with an average age at study enrollment of 56.1 years, and the prevalence of T2D was 36.4%. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, and 10 principal components of Asian genetic ancestry. The MVP genotyped cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8.We performed a European-ancestry single variant analysis including 69,869 type 2 diabetes cases and 127,197 controls in the Million Veteran Program (MVP). Individuals aged 19 to 104 years have been recruited voluntarily from more than 50 VA Medical Centers nationwide for participation in the Million Veteran Program biobank study. Recruitment is currently occurring in person at selected sites in the VHA health care system. European MVP participants are predominantly male (92.8%), with an average age at study enrollment of 68.2 years, and the prevalence of T2D was 35.5%. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with T2D through logistic regression assuming an additive model of variants with MAF > 0.1% using PLINK2a software. Covariates included age, gender, and 10 principal components of European genetic ancestry. The MVP genotyped cohort were imputed using the 1000 Genomes Project phase 3, version 5 reference panel. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8.A Genome-Wide Association Study on PTSD Hyperarousal Symptoms in African Ancestry. A Genome-Wide Association Study on PTSD Hyperarousal Symptoms in European Ancestry. A Genome-Wide Association Study on PTSD Reexperiencing Symptoms in African Ancestry. A Genome-Wide Association Study on PTSD Reexperiencing Symptoms in European Ancestry. A Genome-Wide Association Study on PTSD PCL Symptoms in African Ancestry. A Genome-Wide Association Study on PTSD PCL Symptoms in European Ancestry. A Genome-Wide Association Study on PTSD Avoidance Symptoms in African Ancestry. A Genome-Wide Association Study on PTSD Avoidance Symptoms in European Ancestry. A Genome-Wide Association Study on PTSD Case Control Status in African Ancestry. A Genome-Wide Association Study on PTSD Case Control Status in European Ancestry. A Genome-Wide Association Study on ICD code defined MDD in African Ancestry. A Genome-Wide Association Study on ICD code defined MDD in European Ancestry. A Genome-Wide Association Study on depression symptoms European Ancestry. We performed a African-American restricted meta-analysis including 13,387 proxy NAFLD cases and 23,977 controls in the Million Veteran Program (MVP). Individuals were recruited voluntarily from more than 50 VA Medical Centers nationwide for participation which is still ongoing. The average age at study enrollment was 61.7 years, and prevalence of proxy NAFLD was 41.4%. Genotyping was perfromed on the Affymetrix MVP biobank chip. The cohort was imputed using the 1000 Genomes Project phase 3, version 5 reference panel. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with proxy NAFLD through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, audit-C, and 10 principal components of genetic ancestry. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8.We performed a Asian restricted meta-analysis including 828 proxy NAFLD cases and 1,088 controls in the Million Veteran Program (MVP). Individuals were recruited voluntarily from more than 50 VA Medical Centers nationwide for participation which is still ongoing. The average age at study enrollment was 61.7 years, and prevalence of proxy NAFLD was 41.4%. Genotyping was perfromed on the Affymetrix MVP biobank chip. The cohort was imputed using the 1000 Genomes Project phase 3, version 5 reference panel. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with proxy NAFLD through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, audit-C, and 10 principal components of genetic ancestry. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8.We performed a European-American restricted meta-analysis including 68,725 proxy NAFLD cases and 95,472 controls in the Million Veteran Program (MVP). Individuals were recruited voluntarily from more than 50 VA Medical Centers nationwide for participation which is still ongoing. The average age at study enrollment was 61.7 years, and prevalence of proxy NAFLD was 41.4%. Genotyping was perfromed on the Affymetrix MVP Axiom biobank chip. The cohort was imputed using the 1000 Genomes Project phase 3, version 5 reference panel. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with proxy NAFLD through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, audit-C, and 10 principal components of genetic ancestry. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8 825 Imputation was performed using MiniMac4 and EAGLE, association analysis was performed using PLINK2A.We performed a Hispanic restricted meta-analysis including 7,468 proxy NAFLD cases and 7,650 controls in the Million Veteran Program (MVP). Individuals were recruited voluntarily from more than 50 VA Medical Centers nationwide for participation which is still ongoing. The average age at study enrollment was 61.7 years, and prevalence of proxy NAFLD was 41.4%. Genotyping was perfromed on the Affymetrix MVP biobank chip. The cohort was imputed using the 1000 Genomes Project phase 3, version 5 reference panel. We tested imputed SNPs that passed quality control (e.g. HWE > 1.0x10-10, INFO > 0.3, call rate > 0.975) for association with proxy NAFLD through logistic regression assuming an additive model of variants with MAF > 1% using PLINK2a software. Covariates included age, gender, audit-C, and 10 principal components of genetic ancestry. Variants were considered genome-wide significant if they passed the conventional p-value threshold of 5.0x10-8.We performed a multi-ethnic meta-analysis including 90,408 proxy NAFLD cases and 128,187 controls in the Million Veteran Program (MVP). MVP participants (n = 218595) are comprised predominantly of male subjects (91.6%) and were classified as Europeans (75.1%), African Americans (17.1%), Hispanics (6.9%), and Asians (0.9%). The average age at study enrollment was 61.7 years, and prevalence of proxy NAFLD was 41.4%. Genotyping was performed on the Affymetrix MVP biobank chip. The cohort was imputed using the 1000 Genomes Project phase 3, version 5 reference panel. The prevalence of NAFLD was 41.9% for Europeans, 43.2% for Asians, 49.4% for Hispanics, and 45.8% for African Americans. Only SNPs with ancestry-specific MAF > 1% in these studies were used. We aggregated association summary statistics from the ancestral strata to perform a trans-ancestry meta-analysis. The association summary statistics for each analysis were meta-analyzed in a fixed-effects model using METAL with inverse-variance weighting of log odds ratios. Between-study allelic effect size heterogeneity was assessed with Cochran's Q statistic as implemented in METAL. Variants were considered genome-wide significant if they passed the trans-ancestry p-value threshold of 5.0x10-9. We conducted GWAS and genome-wide meta-analysis of AUD in 13,551 subjects with East Asian ancestry, using published summary data and newly genotyped data from four cohorts: 1) electronic health record (EHR)-diagnosed AUD in the Million Veteran Program (MVP) sample; 2) DSM-IV diagnosed alcohol dependence (AD) in a Han Chinese-GSA (array) cohort; 3) AD in a Han Chinese-Cyto (array) cohort; and 4) two AD datasets in a Thai cohort.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed to a hybrid imputation panel comprised of the African Genome Resources panel (https://imputation.sanger.ac.uk/?about=1#referencepanels) and 1000Genomes (p3v5). King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to three mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, and (3) Hispanic or Latino. We included MVP participants who had genetic data and EHR-extracted COVID-19 related phenotype data available and who were alive as of February 29, 2020. For each pair of relatives (kinship coefficient >= 0.0884), one individual was excluded, preferentially retaining those who tested positive for SARS-CoV-2. Cases of COVID-19 among MVP participants were identified using an algorithm developed by the VA, the COVID National Surveillance Tool (NST) (Chapman et al., 2020). COVID-19-related hospitalizations were defined as admissions from 7 days before up to 30 days after a patient's first positive test for SARS-CoV-2. COVID-19 status was determined between March 1, 2020 and February 2, 2021, a timeframe that represents the first year of the pandemic in the US prior to widespread access to SARS-CoV-2 vaccines and the Delta variant sweep.We performed a genome-wide association study (GWAS) in individuals of non-Hispanic White, non-Hispanic Black, and Hispanic-American race/ethnicity groups in MVP followed by trans-ethnic meta-analysis. GWAS in 63,898 African American individuals did not identify any novel height-associated loci compared to published European-ancestry GWAS. We performed a genome-wide association study (GWAS) in individuals of non-Hispanic White, non-Hispanic Black, and Hispanic-American race/ethnicity groups in MVP followed by trans-ethnic meta-analysis. GWAS in 235,398 European American individuals identified 17 height-associated loci with p<5e-8 that had not previously been reported in European-ancestry height GWAS. We performed a genome-wide association study (GWAS) in individuals of non-Hispanic White, non-Hispanic Black, and Hispanic-American race/ethnicity groups in MVP followed by trans-ethnic meta-analysis. GWAS in 24,497 Hispanic-American individuals did not identify any novel height-associated loci compared to published European-ancestry GWAS. The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. MVP participants were linked to the VA Clinical Assessment, Reporting, and Tracking (CART) Program, a national quality and safety organization for invasive cardiac procedures, to reliably estimate the burden of atherosclerosis among participants who had undergone at least one coronary angiogram by October 2018. Data were available retrospectively starting in 2004 in select sites and from all sites by 2010. For each angiogram, we classified an individual's extent of disease to one of the following categories of disease of the native vessels: normal, non-obstructive, 1 vessel, 2 vessel, 3 vessel and/or left main coronary artery disease. Obstructive disease of a native vessel was defined as the presence of at least one lesion >50% or a prior revascularization procedure involving that vessel. Non-obstructive disease of a native vessel was defined as a vessel with at least one stenosis >20% of luminal diameter but no lesion >50%. We modified a previously validated algorithm to derive these classifications by decreasing the threshold of significant disease in a vessel from at least one lesion >70% to one lesion >50%. Entries were filtered to remove those where disease severity was missing or listed as other, then subjects were removed if they were missing a HARE assignment, date of birth, sex, or had previously received a cardiac transplant. For subjects with multiple angiograms over follow up where at least one reported disease, we assigned severity based on the procedure reporting the most advanced disease. If more than one angiogram reported the same advanced disease, we used the earliest one. Age was calculated on the date of the cardiac catheterization with the most severe disease for cases and the last normal angiogram for controls.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. MVP participants were linked to the VA Clinical Assessment, Reporting, and Tracking (CART) Program, a national quality and safety organization for invasive cardiac procedures, to reliably estimate the burden of atherosclerosis among participants who had undergone at least one coronary angiogram by October 2018. Data were available retrospectively starting in 2004 in select sites and from all sites by 2010. For each angiogram, we classified an individual's extent of disease to one of the following categories of disease of the native vessels: normal, non-obstructive, 1 vessel, 2 vessel, 3 vessel and/or left main coronary artery disease. Obstructive disease of a native vessel was defined as the presence of at least one lesion >50% or a prior revascularization procedure involving that vessel. Non-obstructive disease of a native vessel was defined as a vessel with at least one stenosis >20% of luminal diameter but no lesion >50%. We modified a previously validated algorithm to derive these classifications by decreasing the threshold of significant disease in a vessel from at least one lesion >70% to one lesion >50%. Entries were filtered to remove those where disease severity was missing or listed as other, then subjects were removed if they were missing a HARE assignment, date of birth, sex, or had previously received a cardiac transplant. For subjects with multiple angiograms over follow up where at least one reported disease, we assigned severity based on the procedure reporting the most advanced disease. If more than one angiogram reported the same advanced disease, we used the earliest one. Age was calculated on the date of the cardiac catheterization with the most severe disease for cases and the last normal angiogram for controls.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. MVP participants were linked to the VA Clinical Assessment, Reporting, and Tracking (CART) Program, a national quality and safety organization for invasive cardiac procedures, to reliably estimate the burden of atherosclerosis among participants who had undergone at least one coronary angiogram by October 2018. Data were available retrospectively starting in 2004 in select sites and from all sites by 2010. For each angiogram, we classified an individual's extent of disease to one of the following categories of disease of the native vessels: normal, non-obstructive, 1 vessel, 2 vessel, 3 vessel and/or left main coronary artery disease. Obstructive disease of a native vessel was defined as the presence of at least one lesion >50% or a prior revascularization procedure involving that vessel. Non-obstructive disease of a native vessel was defined as a vessel with at least one stenosis >20% of luminal diameter but no lesion >50%. We modified a previously validated algorithm to derive these classifications by decreasing the threshold of significant disease in a vessel from at least one lesion >70% to one lesion >50%. Entries were filtered to remove those where disease severity was missing or listed as other, then subjects were removed if they were missing a HARE assignment, date of birth, sex, or had previously received a cardiac transplant. For subjects with multiple angiograms over follow up where at least one reported disease, we assigned severity based on the procedure reporting the most advanced disease. If more than one angiogram reported the same advanced disease, we used the earliest one. Age was calculated on the date of the cardiac catheterization with the most severe disease for cases and the last normal angiogram for controls.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. MVP participants were linked to the VA Clinical Assessment, Reporting, and Tracking (CART) Program, a national quality and safety organization for invasive cardiac procedures, to reliably estimate the burden of atherosclerosis among participants who had undergone at least one coronary angiogram by October 2018. Data were available retrospectively starting in 2004 in select sites and from all sites by 2010. For each angiogram, we classified an individual's extent of disease to one of the following categories of disease of the native vessels: normal, non-obstructive, 1 vessel, 2 vessel, 3 vessel and/or left main coronary artery disease. Obstructive disease of a native vessel was defined as the presence of at least one lesion >50% or a prior revascularization procedure involving that vessel. Non-obstructive disease of a native vessel was defined as a vessel with at least one stenosis >20% of luminal diameter but no lesion >50%. We modified a previously validated algorithm to derive these classifications by decreasing the threshold of significant disease in a vessel from at least one lesion >70% to one lesion >50%. Entries were filtered to remove those where disease severity was missing or listed as other, then subjects were removed if they were missing a HARE assignment, date of birth, sex, or had previously received a cardiac transplant. For subjects with multiple angiograms over follow up where at least one reported disease, we assigned severity based on the procedure reporting the most advanced disease. If more than one angiogram reported the same advanced disease, we used the earliest one. Age was calculated on the date of the cardiac catheterization with the most severe disease for cases and the last normal angiogram for controls.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. Inpatient and outpatient ICD diagnostic and CPT procedure codes to identify subjects with clinical CAD in MVP. EHR data was available retrospectively before enrollment going back to October 1999 and prospectively after enrollment until mid-August 2018. An individual was classified as a case if he or she had: 1) any admission to a VA hospital with a discharge diagnosis of acute myocardial infarction (AMI) or 2) any procedure code for revascularization of the coronary arteries, or 3) two or more ICD codes for CAD (410 to 414) in at least two different encounters. Individuals with only one ICD code for CAD in a single encounter and no discharge diagnoses for AMI or revascularization procedures were excluded from the analyses. The remaining subjects were classified as controls. The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. Inpatient and outpatient ICD diagnostic and CPT procedure codes to identify subjects with clinical CAD in MVP. EHR data was available retrospectively before enrollment going back to October 1999 and prospectively after enrollment until mid-August 2018. An individual was classified as a case if he or she had: 1) any admission to a VA hospital with a discharge diagnosis of acute myocardial infarction (AMI) or 2) any procedure code for revascularization of the coronary arteries, or 3) two or more ICD codes for CAD (410 to 414) in at least two different encounters. Individuals with only one ICD code for CAD in a single encounter and no discharge diagnoses for AMI or revascularization procedures were excluded from the analyses. The remaining subjects were classified as controls.The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. Inpatient and outpatient ICD diagnostic and CPT procedure codes to identify subjects with clinical CAD in MVP. EHR data was available retrospectively before enrollment going back to October 1999 and prospectively after enrollment until mid-August 2018. An individual was classified as a case if he or she had: 1) any admission to a VA hospital with a discharge diagnosis of acute myocardial infarction (AMI) or 2) any procedure code for revascularization of the coronary arteries, or 3) two or more ICD codes for CAD (410 to 414) in at least two different encounters. Individuals with only one ICD code for CAD in a single encounter and no discharge diagnoses for AMI or revascularization procedures were excluded from the analyses. The remaining subjects were classified as controls. The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. Inpatient and outpatient ICD diagnostic and CPT procedure codes to identify subjects with clinical CAD in MVP. EHR data was available retrospectively before enrollment going back to October 1999 and prospectively after enrollment until mid-August 2018. An individual was classified as a case if he or she had: 1) any admission to a VA hospital with a discharge diagnosis of acute myocardial infarction (AMI) or 2) any procedure code for revascularization of the coronary arteries, or 3) two or more ICD codes for CAD (410 to 414) in at least two different encounters. Individuals with only one ICD code for CAD in a single encounter and no discharge diagnoses for AMI or revascularization procedures were excluded from the analyses. The remaining subjects were classified as controls. The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. Inpatient and outpatient ICD diagnostic and CPT procedure codes to identify subjects with clinical CAD in MVP. EHR data was available retrospectively before enrollment going back to October 1999 and prospectively after enrollment until mid-August 2018. An individual was classified as a case if he or she had: 1) any admission to a VA hospital with a discharge diagnosis of acute myocardial infarction (AMI) or 2) any procedure code for revascularization of the coronary arteries, or 3) two or more ICD codes for CAD (410 to 414) in at least two different encounters. Individuals with only one ICD code for CAD in a single encounter and no discharge diagnoses for AMI or revascularization procedures were excluded from the analyses. The remaining subjects were classified as controls. The VA Million Veteran Program (MVP) is an ongoing longitudinal study that began in 2011 to study genetic and non-genetic determinants of health and disease among U.S. Veterans. Study participants were genotyped using a customized Affymetrix Axiom biobank array and imputation was performed using the 1000Genomes (p3v5) reference panel. King v2.0 was used to infer kinship and the harmonized race/ethnicity and genetic ancestry (HARE) approach was used to assign individuals to four mutually exclusive groups: (1) non-Hispanic White, (2) non-Hispanic Black, (3) Hispanic or Latino, and (4) Asian. Inpatient and outpatient ICD diagnostic and CPT procedure codes to identify subjects with clinical CAD in MVP. EHR data was available retrospectively before enrollment going back to October 1999 and prospectively after enrollment until mid-August 2018. An individual was classified as a case if he or she had: 1) any admission to a VA hospital with a discharge diagnosis of acute myocardial infarction (AMI) or 2) any procedure code for revascularization of the coronary arteries, or 3) two or more ICD codes for CAD (410 to 414) in at least two different encounters. Individuals with only one ICD code for CAD in a single encounter and no discharge diagnoses for AMI or revascularization procedures were excluded from the analyses. The remaining subjects were classified as controls. Meta-analysis of clinical coronary artery disease in White participants from MVP, UK Biobank, and Cardiogram+C4D.X Chromosome meta-analysis of clinical coronary artery disease in participants from MVP, UK Biobank, Cardiogram+C4D, and Biobank Japan using metalAutosomal meta-analysis of clinical coronary artery disease in participants from MVP, UK Biobank, Cardiogram+C4D, and Biobank Japan using metalMeta-analysis of clinical coronary artery disease in participants from MVP, UK Biobank, Cardiogram+C4D, and Biobank Japan using MR-MEGAOsteoarthritis is a common progressive joint disease. As no effective medical interventions are available, osteoarthritis often progresses to end-stage when only surgical options are available such as total joint replacements. A more thorough understanding of genetic influences of osteoarthritisis essential to identify targeted personalized approaches to treatment, ideally long before end-stage is reached. To date there have been no large multi ancestry genetic studies of osteoarthritis. We leveraged the unique resources of 484,374 participants in the Million Veteran Program (MVP) and UK Biobank to address this gap. Analyses included participants of European, African, Asian and Hispanic descent. We discovered osteoarthritis associated genetic variation in 10 loci and replicated association findings from previous osteoarthritis studies. We also present evidence some osteoarthritis-associated regions are robust to population ancestry. Drug repurposing analyses revealed enrichment of targets of several medication classes and provide potential insight to etiology of beneficial effects of antiepileptics on osteoarthritis pain.Osteoarthritis is a common progressive joint disease. As no effective medical interventions are available, osteoarthritis often progresses to end-stage when only surgical options are available such as total joint replacements. A more thorough understanding of genetic influences of osteoarthritisis essential to identify targeted personalized approaches to treatment, ideally long before end-stage is reached. To date there have been no large multi ancestry genetic studies of osteoarthritis. We leveraged the unique resources of 484,374 participants in the Million Veteran Program (MVP) and UK Biobank to address this gap. Analyses included participants of European, African, Asian and Hispanic descent. We discovered osteoarthritis associated genetic variation in 10 loci and replicated association findings from previous osteoarthritis studies. We also present evidence some osteoarthritis-associated regions are robust to population ancestry. Drug repurposing analyses revealed enrichment of targets of several medication classes and provide potential insight to etiology of beneficial effects of antiepileptics on osteoarthritis pain.Osteoarthritis is a common progressive joint disease. As no effective medical interventions are available, osteoarthritis often progresses to end-stage when only surgical options are available such as total joint replacements. A more thorough understanding of genetic influences of osteoarthritisis essential to identify targeted personalized approaches to treatment, ideally long before end-stage is reached. To date there have been no large multi ancestry genetic studies of osteoarthritis. We leveraged the unique resources of 484,374 participants in the Million Veteran Program (MVP) and UK Biobank to address this gap. Analyses included participants of European, African, Asian and Hispanic descent. We discovered osteoarthritis associated genetic variation in 10 loci and replicated association findings from previous osteoarthritis studies. We also present evidence some osteoarthritis-associated regions are robust to population ancestry. Drug repurposing analyses revealed enrichment of targets of several medication classes and provide potential insight to etiology of beneficial effects of antiepileptics on osteoarthritis pain.Osteoarthritis is a common progressive joint disease. As no effective medical interventions are available, osteoarthritis often progresses to end-stage when only surgical options are available such as total joint replacements. A more thorough understanding of genetic influences of osteoarthritisis essential to identify targeted personalized approaches to treatment, ideally long before end-stage is reached. To date there have been no large multi ancestry genetic studies of osteoarthritis. We leveraged the unique resources of 484,374 participants in the Million Veteran Program (MVP) and UK Biobank to address this gap. Analyses included participants of European, African, Asian and Hispanic descent. We discovered osteoarthritis associated genetic variation in 10 loci and replicated association findings from previous osteoarthritis studies. We also present evidence some osteoarthritis-associated regions are robust to population ancestry. Drug repurposing analyses revealed enrichment of targets of several medication classes and provide potential insight to etiology of beneficial effects of antiepileptics on osteoarthritis pain.Osteoarthritis is a common progressive joint disease. As no effective medical interventions are available, osteoarthritis often progresses to end-stage when only surgical options are available such as total joint replacements. A more thorough understanding of genetic influences of osteoarthritisis essential to identify targeted personalized approaches to treatment, ideally long before end-stage is reached. To date there have been no large multi ancestry genetic studies of osteoarthritis. We leveraged the unique resources of 484,374 participants in the Million Veteran Program (MVP) and UK Biobank to address this gap. Analyses included participants of European, African, Asian and Hispanic descent. We discovered osteoarthritis associated genetic variation in 10 loci and replicated association findings from previous osteoarthritis studies. We also present evidence some osteoarthritis-associated regions are robust to population ancestry. Drug repurposing analyses revealed enrichment of targets of several medication classes and provide potential insight to etiology of beneficial effects of antiepileptics on osteoarthritis pain.We performed genome-wide association study (GWAS) of Suicidal Thoughts and Behaviors (SITB) in the Million Veteran Program (MVP). Total 633,778 U.S. military veterans were identified from electronic health records, included SITB cases 121,211. GWAS was performed separately in four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Cross-ancestry risk loci were identified through meta-analysis.We performed genome-wide association study (GWAS) of Suicidal Thoughts and Behaviors (SITB) in the Million Veteran Program (MVP). Total 633,778 U.S. military veterans were identified from electronic health records, included SITB cases 121,211. GWAS was performed separately in four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Cross-ancestry risk loci were identified through meta-analysis.We performed genome-wide association study (GWAS) of Suicidal Thoughts and Behaviors (SITB) in the Million Veteran Program (MVP). Total 633,778 U.S. military veterans were identified from electronic health records, included SITB cases 121,211. GWAS was performed separately in four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Cross-ancestry risk loci were identified through meta-analysis.We performed genome-wide association study (GWAS) of Suicidal Thoughts and Behaviors (SITB) in the Million Veteran Program (MVP). Total 633,778 U.S. military veterans were identified from electronic health records, included SITB cases 121,211. GWAS was performed separately in four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Cross-ancestry risk loci were identified through meta-analysis.We performed genome-wide association study (GWAS) of Suicidal Thoughts and Behaviors (SITB) in the Million Veteran Program (MVP). Total 633,778 U.S. military veterans were identified from electronic health records, included SITB cases 121,211. GWAS was performed separately in four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Cross-ancestry risk loci were identified through meta-analysis.GWAS on all cause heart failure (HF) of 302287 veterans from the Million Veteran Program with European ancestry. Genetic data was imputed using 1000 Genomes phase 3 panel. HF cases were coded as 2 and controls 1.GWAS on heart failure with preserved ejection fraction (HFpEF) of 278532 veterans from the Million Veteran Program with European ancestry. Genetic data was imputed using 1000 Genomes phase 3 panel. HFpEF cases were coded as 2 and controls 1.GWAS on heart failure with reduced ejection fraction (HFrEF) of 278438 veterans from the Million Veteran Program with European ancestry. Genetic data was imputed using 1000 Genomes phase 3 panel. HFrEF cases were coded as 2 and controls 1.Buprenorphine, approved for treating opioid use disorder (OUD), is not equally efficacious for all patients. We studied 1616 European-ancestry individuals enrolled in the Million Veteran Program, of whom 1609 had an ICD-9/10 code consistent with OUD, a 180-day buprenorphine treatment exposure, and genome-wide genotype data. We conducted a genome-wide association study (GWAS) of buprenorphine treatment response [defined as having no opioid-positive urine drug screens (UDS) following the first prescription]. Imputed genotype data was used. We performed a T2D-related complications analysis in white participants of the Million Veteran Program (MVP) with genetic variants with MAF >5% using SUGEN software with robust standard errors. Within this model, CHD was set as the dependent variable, imputed SNP and type 2 diabetes as the independent variables, together with their interaction term, SNPxT2D. Covariates included age, gender, and 10 principal components of genetic ancestry. Due to privacy concers we only report the beta and p-value for the interaction coefficient. In addition, we provide relative excess risk due to interaction (RERI) for each interaction. We performed a T2D-related complications analysis in white participants of the Million Veteran Program (MVP) with genetic variants with MAF >5% using SUGEN software with robust standard errors. Within this model, CKD was set as the dependent variable, imputed SNP and type 2 diabetes as the independent variables, together with their interaction term, SNPxT2D. Covariates included age, gender, and 10 principal components of genetic ancestry. Due to privacy concers we only report the beta and p-value for the interaction coefficient. In addition, we provide relative excess risk due to interaction (RERI) for each interaction. We performed a T2D-related complications analysis in white participants of the Million Veteran Program (MVP) with genetic variants with MAF >5% using SUGEN software with robust standard errors. Within this model, Retinopathy was set as the dependent variable, imputed SNP and type 2 diabetes as the independent variables, together with their interaction term, SNPxT2D. Covariates included age, gender, and 10 principal components of genetic ancestry. Due to privacy concers we only report the beta and p-value for the interaction coefficient. In addition, we provide relative excess risk due to interaction (RERI) for each interaction. We performed a T2D-related complications analysis in white participants of the Million Veteran Program (MVP) with genetic variants with MAF >5% using SUGEN software with robust standard errors. Within this model, ischemic stroke was set as the dependent variable, imputed SNP and type 2 diabetes as the independent variables, together with their interaction term, SNPxT2D. Covariates included age, gender, and 10 principal components of genetic ancestry. Due to privacy concers we only report the beta and p-value for the interaction coefficient. In addition, we provide relative excess risk due to interaction (RERI) for each interactionWe performed a T2D-related complications analysis in white participants of the Million Veteran Program (MVP) with genetic variants with MAF >5% using SUGEN software with robust standard errors. Within this model, PAD was set as the dependent variable, imputed SNP and type 2 diabetes as the independent variables, together with their interaction term, SNPxT2D. Covariates included age, gender, and 10 principal components of genetic ancestry. Due to privacy concers we only report the beta and p-value for the interaction coefficient. In addition, we provide relative excess risk due to interaction (RERI) for each interaction. We performed a T2D-related complications analysis in white participants of the Million Veteran Program (MVP) with genetic variants with MAF >5% using SUGEN software with robust standard errors. Within this model, Neuropathy was set as the dependent variable, imputed SNP and type 2 diabetes as the independent variables, together with their interaction term, SNPxT2D. Covariates included age, gender, and 10 principal components of genetic ancestry. Due to privacy concers we only report the beta and p-value for the interaction coefficient. In addition, we provide relative excess risk due to interaction (RERI) for each interaction.An ideal health score (IHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels) and 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Binary ideal health (BIH) defined cases as those with an IHS>= 9, and controls with IHS<9. Multi-population genome-wide association study (GWAS) was performed on IHS and BIH using linear and logistic regression, respectively.An ideal health score (IHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels) and 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Binary ideal health (BIH) defined cases as those with an IHS>= 9, and controls with IHS<9. Multi-population genome-wide association study (GWAS) was performed on IHS and BIH using linear and logistic regression, respectively. An ideal health score (IHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels) and 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Binary ideal health (BIH) defined cases as those with an IHS>= 9, and controls with IHS<9. Multi-population genome-wide association study (GWAS) was performed on IHS and BIH using linear and logistic regression, respectively. An ideal health score (IHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels) and 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Binary ideal health (BIH) defined cases as those with an IHS>= 9, and controls with IHS<9. Multi-population genome-wide association study (GWAS) was performed on IHS and BIH using linear and logistic regression, respectively. Association testing was performed in TAAD cases and controls in the Million Veteran Program. Pariticpants of African ancestry were analyzed.Association testing was performed in TAAD cases and controls in the Million Veteran Program. Pariticpants of European ancestry were analyzed.Association testing was performed in TAAD cases and controls in the Million Veteran Program. Pariticpants of Hispanic ancestry were analyzed.Association testing was performed in TAAD cases and controls in the Million Veteran Program. Pariticpants of European (white), African (black), and Hispanic ancestry were stratified by ethnicity followed by a meta-analysis of results across all three groups.A clinical ideal health score (CIHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels), ascertained at baseline. Clinical binary ideal health (CBIH) defined cases as those with a CIHS>= 5, and controls with CIHS<5. Multi-population genome-wide association study (GWAS) was performed on CIHS and CBIH using linear and logistic regression, respectively. A clinical ideal health score (CIHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels), ascertained at baseline. Clinical binary ideal health (CBIH) defined cases as those with a CIHS>= 5, and controls with CIHS<5. Multi-population genome-wide association study (GWAS) was performed on CIHS and CBIH using linear and logistic regression, respectively. A clinical ideal health score (CIHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels), ascertained at baseline. Clinical binary ideal health (CBIH) defined cases as those with a CIHS>= 5, and controls with CIHS<5. Multi-population genome-wide association study (GWAS) was performed on CIHS and CBIH using linear and logistic regression, respectively. A clinical ideal health score (CIHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels), ascertained at baseline. Clinical binary ideal health (CBIH) defined cases as those with a CIHS>= 5, and controls with CIHS<5. Multi-population genome-wide association study (GWAS) was performed on CIHS and CBIH using linear and logistic regression, respectively. A behavioural ideal health score (BIHS) was calculated from 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Behavioural binary ideal health (BBIH) defined cases as those with a BIHS>= 4, and controls with BIHS<4. Multi-population genome-wide association study (GWAS) was performed on BIHS and BBIH using linear and logistic regression, respectively. A behavioural ideal health score (BIHS) was calculated from 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Behavioural binary ideal health (BBIH) defined cases as those with a BIHS>= 4, and controls with BIHS<4. Multi-population genome-wide association study (GWAS) was performed on BIHS and BBIH using linear and logistic regression, respectively. A behavioural ideal health score (BIHS) was calculated from 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Behavioural binary ideal health (BBIH) defined cases as those with a BIHS>= 4, and controls with BIHS<4. Multi-population genome-wide association study (GWAS) was performed on BIHS and BBIH using linear and logistic regression, respectively. A behavioural ideal health score (BIHS) was calculated from 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Behavioural binary ideal health (BBIH) defined cases as those with a BIHS>= 4, and controls with BIHS<4. Multi-population genome-wide association study (GWAS) was performed on BIHS and BBIH using linear and logistic regression, respectively.A behavioural ideal health score (BIHS) was calculated from 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Behavioural binary ideal health (BBIH) defined cases as those with a BIHS>= 4, and controls with BIHS<4. Multi-population genome-wide association study (GWAS) was performed on BIHS and BBIH using linear and logistic regression, respectively. A behavioural ideal health score (BIHS) was calculated from 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Behavioural binary ideal health (BBIH) defined cases as those with a BIHS>= 4, and controls with BIHS<4. Multi-population genome-wide association study (GWAS) was performed on BIHS and BBIH using linear and logistic regression, respectively.An ideal health score (IHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels) and 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Binary ideal health (BIH) defined cases as those with an IHS>= 9, and controls with IHS<9. Multi-population genome-wide association study (GWAS) was performed on IHS and BIH using linear and logistic regression, respectively.An ideal health score (IHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels) and 3 behavioral factors (smoking status, physical activity, and BMI), ascertained at baseline. Binary ideal health (BIH) defined cases as those with an IHS>= 9, and controls with IHS<9. Multi-population genome-wide association study (GWAS) was performed on IHS and BIH using linear and logistic regression, respectively. A clinical ideal health score (CIHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels), ascertained at baseline. Clinical binary ideal health (CBIH) defined cases as those with a CIHS>= 5, and controls with CIHS<5. Multi-population genome-wide association study (GWAS) was performed on CIHS and CBIH using linear and logistic regression, respectively. A clinical ideal health score (CIHS) was calculated from 3 clinical factors (blood pressure, total cholesterol, and blood glucose levels), ascertained at baseline. Clinical binary ideal health (CBIH) defined cases as those with a CIHS>= 5, and controls with CIHS<5. Multi-population genome-wide association study (GWAS) was performed on CIHS and CBIH using linear and logistic regression, respectively.We performed the first GWAS of suicidal ideation (SI) without suicide attempt (SA) in the Million Veteran Program (MVP), identifying 99,814 SI cases from electronic health records without a history of SA or suicide death (SD) and 512,567 controls without SI, SA or SD. GWAS was performed separately in the four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Ancestry-specific results were combined via meta-analysis to identify pan-ancestry loci.We performed the first GWAS of suicidal ideation (SI) without suicide attempt (SA) in the Million Veteran Program (MVP), identifying 99,814 SI cases from electronic health records without a history of SA or suicide death (SD) and 512,567 controls without SI, SA or SD. GWAS was performed separately in the four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Ancestry-specific results were combined via meta-analysis to identify pan-ancestry loci.We performed the first GWAS of suicidal ideation (SI) without suicide attempt (SA) in the Million Veteran Program (MVP), identifying 99,814 SI cases from electronic health records without a history of SA or suicide death (SD) and 512,567 controls without SI, SA or SD. GWAS was performed separately in the four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Ancestry-specific results were combined via meta-analysis to identify pan-ancestry loci.We performed the first GWAS of suicidal ideation (SI) without suicide attempt (SA) in the Million Veteran Program (MVP), identifying 99,814 SI cases from electronic health records without a history of SA or suicide death (SD) and 512,567 controls without SI, SA or SD. GWAS was performed separately in the four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Ancestry-specific results were combined via meta-analysis to identify pan-ancestry loci.We performed the first GWAS of suicidal ideation (SI) without suicide attempt (SA) in the Million Veteran Program (MVP), identifying 99,814 SI cases from electronic health records without a history of SA or suicide death (SD) and 512,567 controls without SI, SA or SD. GWAS was performed separately in the four largest ancestry groups (European Ancestry, African Ancestry, Hispanic Ancestry and Asian Ancestry), controlling for sex, age and genetic substructure. Ancestry-specific results were combined via meta-analysis to identify pan-ancestry loci.We performed a GWAS meta-analysis of 45,254 European ancestry individuals with varicose veins and 985,690 disease-free controls, identified from the VA Million Veteran Program, UK Biobank, FinnGen, and eMERGE. Standard genotyping and quality control measures were applied within each cohort. Individuals with varicose veins were identified from electronic health record diagnosis codes (ICD9 454, ICD10 I83, or pheCode 454.1).We performed a GWAS meta-analysis of 1,226 Admixed American/Hispanic individuals with varicose veins and 45,168 disease-free controls, identified from the VA Million Veteran Program. Standard genotyping and quality control measures were applied within each cohort. Individuals with varicose veins were identified from electronic health record diagnosis codes (ICD9 454, ICD10 I83, or pheCode 454.1).We performed a GWAS meta-analysis of 49,765 individuals with varicose veins and 1,334,301 disease-free controls, identified from the VA Million Veteran Program, UK Biobank, FinnGen, BioBank Japan, and eMERGE. Standard genotyping and quality control measures were applied within each cohort. Individuals with varicose veins were identified from electronic health record diagnosis codes (ICD9 454, ICD10 I83, or pheCode 454.1).GWAS of BMI of 55,525 Veterans from the Million Veteran Program with African ancestry. Genetic data was imputed using 1000 Genomes phase 3 panel. GWAS on BMI of 215,734 Veterans from the Million Veteran Program with European ancestry. Genetic data was imputed using 1000 Genomes phase 3 panel. We performed a GWAS meta-analysis of 2,811 African ancestry individuals with varicose veins and 125,191 disease-free controls, identified from the VA Million Veteran Program, UK Biobank, and eMERGE. Standard genotyping and quality control measures were applied within each cohort. Individuals with varicose veins were identified from electronic health record diagnosis codes (ICD9 454, ICD10 I83, or pheCode 454.1).We leverage BED (binge-eating disorder) diagnosis codes based on electronic medical records from the Million Veteran Program. We perform a genome-wide association study on this BED phenotype while controlling for body mass index.We apply a supervised machine learning approach to estimate the probability of each individual having BED (binge-eating disorder) based on electronic medical records from the Million Veteran Program. We perform a genome-wide association study on this model derived BED probability phenotype while controlling for body mass index.We apply a supervised machine learning approach to estimate the probability of each individual having BED (binge-eating disorder) based on electronic medical records from the Million Veteran Program. We perform a genome-wide association study on this model derived BED probability phenotype while controlling for body mass index.These data represent summary statistics from GWAS for calcific aortic stenosis (CAS) in the Million Veteran Program (MVP). MVP is a large biobank consisting of individuals aged 19 years and older in the Veterans Affairs Healthcare system. A CAS phenotype was generated using ICD/CPT codes (see definition in corresponding manuscript) amounting to 1,445 Black cases and 79,229 Black controls. Genotyping took place on a customized Affymetrix Axion Biobank array. Quality control for MVP samples are described in the corresponding manuscript.These data represent summary statistics from GWAS for calcific aortic stenosis (CAS) in the Million Veteran Program (MVP). MVP is a large biobank consisting of individuals aged 19 years and older in the Veterans Affairs Healthcare system. A CAS phenotype was generated using ICD/CPT codes (see definition in corresponding manuscript) amounting to 12,395 White cases and 287,787 White controls. Genotyping took place on a customized Affymetrix Axion Biobank array. Quality control for MVP samples are described in the corresponding manuscript.These data represent summary statistics from GWAS for calcific aortic stenosis (CAS) in the Million Veteran Program (MVP). MVP is a large biobank consisting of individuals aged 19 years and older in the Veterans Affairs Healthcare system. A CAS phenotype was generated using ICD/CPT codes (see definition in corresponding manuscript) amounting to 611 Hispanic cases and 31,458 Hispanic controls. Genotyping took place on a customized Affymetrix Axion Biobank array. Quality control for MVP samples are described in the corresponding manuscript.These data represent summary statistics from meta-analysis of GWAS for calcific aortic stenosis (CAS) in the Million Veteran Program (MVP). MVP is a large biobank consisting of individuals aged 19 years and older in the Veterans Affairs Healthcare system. A CAS phenotype was generated using ICD/CPT codes (see definition in corresponding manuscript) amounting to 14,451 CAS cases and 398,544 controls (12,395 White cases, 287,787 White controls; 1,445 Black cases, 79,229 Black controls; 611 Hispanic cases, 31,458 Hispanic controls). Genotyping took place on a customized Affymetrix Axion Biobank array. Quality control for MVP samples are described in the corresponding manuscript.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed GWAS of OSA in 568,576 MVP participants (cases= 121332 and control=447244). Participants of White, Black, Hispanic/Latino, and Asian were stratified by HARE group and sex, followed by meta-analysis.We performed a GWAS meta-analysis comprised of five European-ancestry cohorts of age-related macular degeneration (AMD) cases and controls: three tranches of MVP EUR, IAMDGC, GERA, UKB, Genentech Geographic Atrophy, and Genentech Choroidal Neovascularization.We performed a multi-ancestry meta-analysis of age-related macular degeneration (AMD) cases and controls. This meta-analysis was comprised of five European-ancestry cohorts (three tranches of MVP EUR, IAMDGC, GERA, UK Biobank, Genentech Geographic Atrophy, and Genentech Choroidal Neovascularization), one African-ancestry cohort (MVP AFR), and one Hispanic/Latino-ancestry cohort (MVP HIS).GWAS of age-related macular degeneration (AMD) were peformed in African ancestry participants (based on Harmonized Ancestry and Race/Ethnicity method; PMID 31564439) in the Million Veteran Program. We used ICD9/ICD10 codes in VA electronic health records to define AMD case/control status, using Algorithm 4 in Halladay et al. (PMID 31258967) in which AMD cases and controls were required to be 50 and 65 years of age, respectively. Briefly, AMD cases had ICD9/ICD10 codes indicating AMD from at least two separate eye clinic visits; controls had no AMD diagnoses on at least two separate eye clinic visits. Samples were genotyped on the ThermoFisher MVP 1.0 Axiom array and imputed to the 1000 Genomes and African Genome Resources reference panel.GWAS of age-related macular degeneration (AMD) cases and controls were peformed in three tranches of European ancestry participants from the Million Veteran Program. We used ICD9/ICD10 codes in VA electronic health records to define AMD case/control status, using Algorithm 4 in Halladay et al. (PMID 31258967) in which AMD cases and controls were required to be 50 and 65 years of age, respectively. Briefly, AMD cases had ICD9/ICD10 codes indicating AMD from at least two separate eye clinic visits; controls had no AMD diagnoses on at least two separate eye clinic visits. The three EA cohorts represent three separate releases of the MVP genotype data. Samples were genotyped on the ThermoFisher MVP 1.0 Axiom array and imputed to the 1000 Genomes reference panel.GWAS of age-related macular degeneration (AMD) were peformed in Hispanic/Latino ancestry participants (based on Harmonized Ancestry and Race/Ethnicity method; PMID 31564439) in the Million Veteran Program. We used ICD9/ICD10 codes in VA electronic health records to define AMD case/control status, using Algorithm 4 in Halladay et al. (PMID 31258967) in which AMD cases and controls were required to be 50 and 65 years of age, respectively. Briefly, AMD cases had ICD9/ICD10 codes indicating AMD from at least two separate eye clinic visits; controls had no AMD diagnoses on at least two separate eye clinic visits. Samples were genotyped on the ThermoFisher MVP 1.0 Axiom array and imputed to the 1000 Genomes and African Genome Resources reference panel.Cases and controls were selected from subjects of at least 65 years of age in the UK Biobank. Samples were genotyped on the UK BiLEVE Axiom Array or the UK Biobank Axiom Array and imputed to the Haplotype Reference Consortium (HRC) reference panel.We performed whole-genome sequencing of DNA derived from blood samples obtained from age-related macular degeneration (AMD) patients with choroidal neovascularization (CNV) participating in clinical trials for Ranibizumab (NCT00891735 [HARBOR], NCT00061594 [ANCHOR] and NCT00056836 [MARINA]). Patients had to consent for genetic analysis for inclusion eligibility and these cohorts were selected for inclusion based on availability of DNA and available phenotypic information. Samples and data for non-AMD controls were obtained from clinical trial cohorts of asthma and RA. The control sample did not overlap with the AMD geographic atrophy (GA) GWAS. All patients (AMD and non-AMD controls) included in this study provided written informed consent for whole-genome sequencing of their DNA. Ethical approval was obtained as per the original clinical trials.We performed whole-genome sequencing of DNA derived from blood samples obtained from age-related macular degeneration (AMD) patients with geographic atrophy (GA) participating in clinical trials for Lampalizumab (NCT02247479 [CHROMA], NCT02247531 [SPECTRI], NCT01229215 [MAHALO]) and an observational study (NCT02479386 [PROXIMA]). Patients had to consent for genetic analysis for inclusion eligibility and these cohorts were selected for inclusion based on availability of DNA and available phenotypic information. Samples and data for non-AMD controls were obtained from clinical trial cohorts of asthma, colorectal cancer, COPD, inflammatory bowel disease, IPF, and RA. The control sample did not overlap with the AMD choroidal neovascularization (CNV) GWAS. All patients (AMD and non-AMD controls) included in this study provided written informed consent for whole-genome sequencing of their DNA. Ethical approval was obtained as per the original clinical trials.GWAS meta-analysis of Fuchs endothelial corneal dystrophy (FECD) was performed across European participants from Million Veteran Program (2,251 cases, 252,345 controls) and a previous GWAS (Afshari et al. 2017; 1,404 cases, 2,564 controls), for a total of 3,655 cases and 254,909 controls. MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh38.GWAS was performed in 2,251 cases and 252,345 controls of European ancestry from Million Veteran Program. Cases were identified based on the presence of FECD codes (371.57 for ICD-9-CM; H18.51 for ICD-10-CM) on two separate visits and the absence of ICD-9-CM or ICD-10-CM codes for confounding corneal conditions or complicated intraocular surgeries. Controls without FECD were identified as having undergone at least one eye exam, with no codes for FECD, confounding corneal conditions, or complicated intraocular surgeries. MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh38.GWAS meta-analysis of Fuchs endothelial corneal dystrophy (FECD) was performed across European participants from Million Veteran Program (2,251 cases, 252,345 controls), African ancestry participants from Million Veteran Program (315 cases, 78,885 controls), and a previous European ancestry GWAS (Afshari et al. 2017; 1,404 cases, 2,564 controls), for a total of 3,970 cases and 333,794 controls. MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh38.We recruited a cohort of 17,213 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We recruited a cohort of 62,232 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We recruited a cohort of 5,289 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We used a cohort of 86,237 participants, with mean age of 69y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We recruited a cohort of 17,213 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We recruited a cohort of 62,232 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We recruited a cohort of 5,289 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array.We recruited a cohort of 86,237 participants, with mean age of 69y, from participants in the Million Veteran Program (MVP) and 7,185 participants, with mean age of 62y, from the UK Biobank (UKB.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array. Imputation was performed using a 1000 Genomes Project Phase 3 v.5 reference panel after pre-phasing with EAGLE, supplemented with the African Genome Resources panel. For individuals in UKB, genotype imputation to a reference set combining the UK10K haplotype and HRC reference panels was performed using IMPUTE2 algorithms.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.Recent genome-wide association studies (GWASs) of alcohol-related phenotypes have uncovered key differences in the underlying genetic architectures of alcohol consumption and alcohol use disorder (AUD), with the two traits having opposite genetic correlations with psychiatric disorders. Understanding the genetic factors that underlie the transition from heavy drinking to AUD has important theoretical and clinical implications. We conducted a GWAS of AUD and alcohol consumption (measured by the score on the consumption subscale of the Alcohol Use Disorders Identification Test [AUDIT-C]) in the MVP in three ancestry groups, followed by a cross ancestry meta-analysis (N=409,630). Cases had received at least one inpatient or two outpatient lifetime ICD Ninth Revision (ICD-9) or Tenth Revision (ICD-10) diagnoses of AUD. Imputed genotype data was used.We conducted GWAS and meta-analyses of alcohol use disorder (AUD) or problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of alcohol use disorder (AUD) or problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of alcohol use disorder (AUD) or problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of alcohol use disorder (AUD) or problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We conducted GWAS and meta-analyses of problematic alcohol use (PAU) using newly genotyped subjects and previously published data from multiple cohorts (MVP, FinnGen, UK Biobank, Psychiatric Genomics Consortium, iPSYCH, QIMR Berghofer Medical Research Institute cohorts, Yale-Penn 3, and East Asian cohorts).We recruited a cohort of 123,024 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array. Imputation was performed using a 1000 Genomes Project Phase 3 v.5 reference panel after pre-phasing with EAGLE, supplemented with the African Genome Resources panel.We recruited a cohort of 464,647 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array. Imputation was performed using a 1000 Genomes Project Phase 3 v.5 reference panel after pre-phasing with EAGLE, supplemented with the African Genome Resources panel.We recruited a cohort of 52,132 participants, with mean age of 65y, from participants in the Million Veteran Program (MVP.) Consented individuals in the MVP were genotyped using a customized Affymetrix Axiom Biobank Array. Imputation was performed using a 1000 Genomes Project Phase 3 v.5 reference panel after pre-phasing with EAGLE, supplemented with the African Genome Resources panel.The purpose of the analysis was to identify risk variants for dementia in African Americans. Samples were genotyped on the MVP 1.0 custom Axiom array. The total sample size was 20,372. The phenotype was Alzheimer's Disease, coded 0 or 1.The purpose of the analysis was to identify risk variants for dementia in African Americans. Samples were genotyped on the MVP 1.0 custom Axiom array. The total sample size was 22,447. The phenotype was ADRD, coded 0 or 1. Cases had an onset age >=60 years. Controls were all age 65 or over at last visit.The purpose of the analysis was to identify risk variants for dementia in African Americans. Samples were genotyped on the MVP 1.0 custom Axiom array. The total sample size was 22,447. The phenotype was ADRD, coded 0 or 1. Cases had an onset age >=60 years. Controls were all age 65 or over at last visit.These results represent the full meta-analysis results described in Sherva et al. 2023 (African ancestry GWAS of dementia in a large military cohort identifies significant risk loci)The purpose of the analysis was to identify risk variants for dementia in African Americans. Samples were genotyped on the MVP 1.0 custom Axiom array. The total sample size was 39,797. The phenotype was maternal dementia, coded 0 or 1.The purpose of the analysis was to identify risk variants for dementia in African Americans. Samples were genotyped on the MVP 1.0 custom Axiom array. The total sample size was 38,130. The phenotype was paternal dementia, coded 0 or 1.Cases were Veterans with an endorsement of concussion or loss of consciousness or traumatic brain injuryon the MVP Baseline Survey; endorsement of a TBI-associated sign or symptom on the MVP Lifestyle Survey; or at least 1 ICD-9 or ICD-10 code according to the TBI definition set the by Armed Forces Health Surveillance Branch. Controls had to complete both the MVP Baseline and Lifestyle Surveys and not reported any event consistent with TBI on either survey. Additionally, controls had to have evidence of VHA usage through ICD-9 and ICD-10 codes, but no TBI ICD codes. Analyses were done from release 4 of the MVP imputed genotype data; refer to Gaziano et al. (2016) and Hunter-Zinck et al. (2020) for details of genotyping, quality control, and imputation procedures. HARE estimates (Fang et al., 2019) defined participants as European, African, and Hispanic ancestry. One individual from each pair of related individuals was removed with preference for retaining cases. Genome-wide association analysis was calculated in PLINK, using logistic regression and 10 principal components calculated within ancestry group as covariates.Cases were Veterans with an endorsement of concussion or loss of consciousness or traumatic brain injury on the MVP Baseline Survey; endorsement of a TBI-associated sign or symptom on the MVP Lifestyle Survey; or at least 1 ICD-9 or ICD-10 code according to the TBI definition set the by Armed Forces Health Surveillance Branch. Controls had to complete both the MVP Baseline and Lifestyle Surveys and not reported any event consistent with TBI on either survey. Additionally, controls had to have evidence of VHA usage through ICD-9 and ICD-10 codes, but no TBI ICD codes. Analyses were done from release 4 of the MVP imputed genotype data; refer to Gaziano et al. (2016) and Hunter-Zinck et al. (2020) for details of genotyping, quality control, and imputation procedures. HARE estimates (Fang et al., 2019) defined participants as European, African, and Hispanic ancestry. One individual from each pair of related individuals was removed with preference for retaining cases. Genome-wide association analysis was calculated in PLINK, using logistic regression and 10 principal components calculated within ancestry group as covariates.Cases were Veterans with an endorsement of concussion or loss of consciousness or traumatic brain injury on the MVP Baseline Survey; endorsement of a TBI-associated sign or symptom on the MVP Lifestyle Survey; or at least 1 ICD-9 or ICD-10 code according to the TBI definition set the by Armed Forces Health Surveillance Branch. Controls had to complete both the MVP Baseline and Lifestyle Surveys and not reported any event consistent with TBI on either survey. Additionally, controls had to have evidence of VHA usage through ICD-9 and ICD-10 codes, but no TBI ICD codes. Analyses were done from release 4 of the MVP imputed genotype data; refer to Gaziano et al. (2016) and Hunter-Zinck et al. (2020) for details of genotyping, quality control, and imputation procedures. HARE estimates (Fang et al., 2019) defined participants as European, African, and Hispanic ancestry. One individual from each pair of related individuals was removed with preference for retaining cases. Genome-wide association analysis was calculated in PLINK, using logistic regression and 10 principal components calculated within ancestry group as covariates.Cases were Veterans with an endorsement of concussion or loss of consciousness or traumatic brain injury on the MVP Baseline Survey; endorsement of a TBI-associated sign or symptom on the MVP Lifestyle Survey; or at least 1 ICD-9 or ICD-10 code according to the TBI definition set the by Armed Forces Health Surveillance Branch. Controls had to complete both the MVP Baseline and Lifestyle Surveys and not reported any event consistent with TBI on either survey. Additionally, controls had to have evidence of VHA usage through ICD-9 and ICD-10 codes, but no TBI ICD codes. Analyses were done from release 4 of the MVP imputed genotype data; refer to Gaziano et al. (2016) and Hunter-Zinck et al. (2020) for details of genotyping, quality control, and imputation procedures. HARE estimates (Fang et al., 2019) defined participants as European, African, and Hispanic ancestry. One individual from each pair of related individuals was removed with preference for retaining cases. Genome-wide association analysis was calculated in PLINK, using logistic regression and 10 principal components calculated within ancestry group as covariates.GWAS of mosaic loss of X (mLOX) was peformed in European ancestry participants (based on Harmonized Ancestry and Race/Ethnicity method; PMID 31564439) in the Million Veteran Program. Mosaic loss of X cases and controls were detepiretinal membrane (ERM)ined using the MoChA software tool (https://github.com/freeseek/mocha). Samples were genotyped on the Thepiretinal membrane (ERM)oFisher MVP 1.0 Axiom array and imputed to the TOPMed reference panel.A Genome-Wide Association Study on ICD code defined epiretinal membrane (ERM) in African Ancestry.A Genome-Wide Association Study on ICD code defined epiretinal membrane (ERM) in Latin American Ancestry.A Genome-Wide Association Study on ICD code defined epiretinal membrane (ERM) in European Ancestry.Chronic pain is a common problem, with more than one-fifth of adult Americans reporting pain daily or on most days. It adversely affects the quality of life and imposes substantial personal and economic costs. Efforts to treat chronic pain using opioids had a central role in precipitating the opioid crisis. Despite an estimated heritability of 25-50%, the genetic architecture of chronic pain is not well-characterized, in part because studies have largely been limited to samples of European ancestry. To help address this knowledge gap, we conducted a cross-ancestry meta-analysis of pain intensity in the Million Veteran Program in three ancestry groups, followed by a cross ancestry meta-analysis among 598,339 participants. The numerical pain rating scale (median of annual median pain score) was used to assess pain intensity. Imputed genotype data was used.Chronic pain is a common problem, with more than one-fifth of adult Americans reporting pain daily or on most days. It adversely affects the quality of life and imposes substantial personal and economic costs. Efforts to treat chronic pain using opioids had a central role in precipitating the opioid crisis. Despite an estimated heritability of 25-50%, the genetic architecture of chronic pain is not well-characterized, in part because studies have largely been limited to samples of European ancestry. To help address this knowledge gap, we conducted a cross-ancestry meta-analysis of pain intensity in the Million Veteran Program in three ancestry groups, followed by a cross ancestry meta-analysis among 598,339 participants. The numerical pain rating scale (median of annual median pain score) was used to assess pain intensity. Imputed genotype data was used.Chronic pain is a common problem, with more than one-fifth of adult Americans reporting pain daily or on most days. It adversely affects the quality of life and imposes substantial personal and economic costs. Efforts to treat chronic pain using opioids had a central role in precipitating the opioid crisis. Despite an estimated heritability of 25-50%, the genetic architecture of chronic pain is not well-characterized, in part because studies have largely been limited to samples of European ancestry. To help address this knowledge gap, we conducted a cross-ancestry meta-analysis of pain intensity in the Million Veteran Program in three ancestry groups, followed by a cross ancestry meta-analysis among 598,339 participants. The numerical pain rating scale (median of annual median pain score) was used to assess pain intensity. Imputed genotype data was used.Chronic pain is a common problem, with more than one-fifth of adult Americans reporting pain daily or on most days. It adversely affects the quality of life and imposes substantial personal and economic costs. Efforts to treat chronic pain using opioids had a central role in precipitating the opioid crisis. Despite an estimated heritability of 25-50%, the genetic architecture of chronic pain is not well-characterized, in part because studies have largely been limited to samples of European ancestry. To help address this knowledge gap, we conducted a cross-ancestry meta-analysis of pain intensity in the Million Veteran Program in three ancestry groups, followed by a cross ancestry meta-analysis among 598,339 participants. The numerical pain rating scale (median of annual median pain score) was used to assess pain intensity. Imputed genotype data was used.Chronic age-related imbalance is a common cause of falls and subsequent death in the elderly and can arise from dysfunction of the vestibular system. Total sample size = 417,239 that had been genotyped in the Million Veteran Program. The phenotype comprised cases with at least 2 ICD diagnoses of vertigo or dizziness at least six months apart, with removal of acute syndromes, cerebellar, cerebrovascular diagnoses, and traumatic brain injury. Genome-wide association studies were performed as separate logistic regressions on European, African American, Asian, and Hispanic ancestries followed by trans-ancestry meta-analysis. minus the Asian population, since there were too few subjects for power. GWAS was calculated for two related phenotypes: two diagnoses >= 180 days apart on LA ancestries (N =34,108, cases = 3,631, controls = 30,481) and >= 90 days apart, identical findings).Chronic age-related imbalance is a common cause of falls and subsequent death in the elderly and can arise from dysfunction of the vestibular system. Total sample size = 417,239 that had been genotyped in the Million Veteran Program. The phenotype comprised cases with at least 2 ICD diagnoses of vertigo or dizziness at least six months apart, with removal of acute syndromes, cerebellar, cerebrovascular diagnoses, and traumatic brain injury. Genome-wide association studies were performed as separate logistic regressions on European, African American, Asian, and Hispanic ancestries followed by trans-ancestry meta-analysis. minus the Asian population, since there were too few subjects for power. GWAS was calculated for two related phenotypes: two diagnoses >= 90 days apart on each ancestry - EU, AA, and LA ancestries (N = 427,959, cases = 55,404, controls = 372,555) and >= 180 days apart.A Genome-Wide Association Study of physical activity during leisure time in European Ancestry. A Genome-Wide Association Study of vigorous physical activity during home time in European Ancestry. A Genome-Wide Association Study of vigorous physical activity during leisure time in European Ancestry. A Genome-Wide Association Study of vigorous physical activity during work time in European Ancestry. GWAS was performed across European ancestry participants of Million Veteran Program (10,398 cases, 62,708 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS was performed across African ancestry participants of Million Veteran Program (2,438 cases, 62,112 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (10,398 cases, 62,708 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 29,266 cases, 56,450  controls), and African ancestry participants of MVP (2,438 cases, 62,112 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (10,398 cases, 62,708 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 29,266 cases, 56,450  controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.Summary statistics from the GWAS meta-analysis  performed across European ancestry participants of Million Veteran Program (10,398 cases, 62,708 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 29,266 cases, 56,450 controls) were conditioned using summary statistics from a GWAS of cigarettes smoked per day (PMID: 30643251). Genomic coordinates correspond to GRCh37.GWAS was performed across European ancestry participants of Million Veteran Program (2,019 cases, 12,113 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS was performed across African ancestry participants of Million Veteran Program (595 cases, 62,112 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (2,019 cases, 12,113 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 11,273 cases, 55,483 controls), and African ancestry participants of MVP (595 cases, 62,112 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (2,019 cases, 12,113 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 11,273 cases, 55,483 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.Summary statistics from the GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (2,019 cases, 12,113 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 11,273 cases, 55,483 controls) were conditioned using summary statistics from a GWAS of cigarettes smoked per day (PMID: 30643251). Genomic coordinates correspond to GRCh37.GWAS was performed across European ancestry participants of Million Veteran Program (1,475 cases, 8,850 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (1,475 cases, 8,850 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 7,426  cases, 55,627 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (1,475 cases, 8,850 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 7,426  cases, 55,627 controls) and African ancestry participants of MVP (300 cases, 62,112 controls). MVP samples were genotyped on the ThermoFisher MVP 1.0 Axiom array. Genomic coordinates correspond to GRCh37.Summary statistics from the GWAS meta-analysis was performed across European ancestry participants of Million Veteran Program (1,475 cases, 8,850 controls) and the International Lung Cancer Consortium OncoArray study (PMID: 28604730; 7,426  cases, 55,627 controls) were conditioned using summary statistics from a GWAS of cigarettes smoked per day (PMID: 30643251). Genomic coordinates correspond to GRCh37.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for acute kidney injury using samples from Million Veteran Program.Contains summary statistics for common variants (MAF > 1%) from discovery GWAS analyses for acute kidney injury using samples from Million Veteran Program.Contains summary statistics for common variants (MAF > 0.1%) from discovery GWAS analyses for acute kidney injury using samples from Million Veteran Program.Contains summary statistics for common variants (MAF > 0.1%) from discovery GWAS analyses for acute kidney injury using samples from Million Veteran Program.Contains summary statistics for common variants (MAF > 0.1%) from discovery GWAS analyses for acute kidney injury using samples from Million Veteran Program and BioVU.Contains summary statistics for common variants (MAF > 0.1%) from discovery GWAS analyses for acute kidney injury using samples from Million Veteran Program and BioVU.Contains summary statistics  GWAS analyses for chronic back pain using samples from Million Veteran Program.Contains summary statistics  GWAS analyses for chronic back pain using samples from Million Veteran Program.Contains summary statistics  GWAS analyses for chronic back pain using samples from Million Veteran Program.Contains summary statistics  GWAS analyses for chronic back pain using samples from Million Veteran Program.Contains summary statistics  GWAS analyses for chronic back pain using samples from Million Veteran Program.Contains summary statistics GWAS analyses for chronic back pain using samples from Million Veteran Program.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.A total of 61,043 MVP participants (3.7% female; 74.2% EUR, 16.9% AFR, 7.3% AMR, 0.6% EAS, 0.05% SAS, and 0.95% other) with at least one coronary angiogram evaluation that included a documentation of coronary artery dominance pattern in the CART registry. Coronary artery dominance is collected as a structured variable in the CART reporting template report used to document the results of coronary angiogram and is typically entered by a member of the team who performed or witnessed the procedure. One of three dominance options can be selected including right, co-dominant, and left. Association testing was performed in MVP participants with coronary dominance stratified by genetically inferred ancestry and with all subjects combined.We performed meta-analysis of GWAS for bradyarrhythmias from up to 1.3M individuals from multiple studies. We performed meta-analysis of GWAS for bradyarrhythmias from up to 1.3M individuals from multiple studies. We performed meta-analysis of GWAS for bradyarrhythmias from up to 1.3M individuals from multiple studies. We performed meta-analysis of GWAS for bradyarrhythmias from up to 1.3M individuals from multiple studies.We performed meta-analysis of GWAS for bradyarrhythmias from up to 1.3M individuals from multiple studies.We performed meta-analysis of GWAS for supraventricular arrhythmias from up to 1.48M individuals from multiple studies. We performed meta-analysis of GWAS for supraventricular arrhythmias from up to 1.48M individuals from multiple studies.We performed meta-analysis of GWAS for supraventricular arrhythmias from up to 1.48M individuals from multiple studies.We performed meta-analysis of GWAS for supraventricular arrhythmias from up to 1.48M individuals from multiple studies.Genomewide association study (GWAS) of dilated cardiomyopathy in African American participants from the VA Million Veteran Program (MVP).Eligible subjects must be active users of the Veteran Health Administration (with available electronic health record) and willing to provide informed consent.]]>
We began enrollment in 2011, initially at VA facilities in Boston, Massachusetts, and West Haven, Connecticut. Subsequently, other recruitment centers were added nationwide. All participants provided informed consent, and the study protocol was approved by the Veterans Affairs Central Institutional Review Board.]]>